A multi-analytical approach for the identification of surface whitening phenomena in contemporary oil painting and its application to metal soaps

Journal of Cultural Heritage 74 (2025) 195–203 Contents lists available at ScienceDirect Journal of Cultural Heritage journal homepage: www.elsevier.com/locate/culher VSI: TechnoHeritage2024 A multi-analytical approach for the identification of surface whitening phenomena in contemporary oil painting and its application to metal soaps Erika Tarilonte a , Oskar González-Mendia a , ∗,Ilaria Costantini b ,Kepa Castro b , Itxaso Maguregui a a Department of Painting, University of the Basque Country UPV/EHU, Leioa, Bizkaia, Spain b IBeA Research Group, University of the Basque Country UPV/EHU, Leioa, Bizkaia, Spain a r t i c l e i n f o Article history: Received 20 February 2025 Accepted 10 June 2025 Available online 27 June 2025 Keywords: White haze Metal soaps Oil painting Multi-analytical study a b s t r a c t This research presents a multi-analytical approach for the characterization of the degradation products that form white hazes on paint surfaces. The surface of Untitled (1971), an oil painting by Santos Iñurrieta (1950–2023), one of the most noteworthy contemporary Basque artists, was chosen as a representative example. The appearance of surface whitening phenomena is a major issue in oil painting conservation and their identification represents a challenge since their origin can be very diverse. In order to characterize the white substance over Untitled (1971) and understand its formation, a global examination was performed, including radiography and UV induced fluorescence, followed by the analysis of different types of microsamples according to the diagnostic technique applied: paint fragments, cross-sections or surface scrapings. The methodology included the application of several techniques on the aforementioned samples: Digital Microscopy (DM), Optical Microscopy (OM), Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEM-EDX), X-ray Computed Tomography (CT), micro X-Ray Fluorescence (μED-XRF), Raman Spectroscopy, Attenuated Total Reflection-Fourier Transform Infrared spectroscopy (ATRFTIR), X-ray Diffraction (XRD), Gas Chromatography-Mass Spectrometry (GC–MS) and X-ray Photoelectron Spectroscopy (XPS). The proposed approach resulted in the identification of lead soaps (palmitate and stearate) as the primary cause for the whitening of the painting surface. Additionally, it provided information of the materials and techniques employed by the artist and revealed that the lead source was probably a lead drier. Thus, this work is a step forward in order to deal with one of the most significant problems currently being addressed in contemporary art conservation, providing relevant information for the efficient removal of the different whitening phenomena. ©2025 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY-NC license ( http://creativecommons.org/licenses/by-nc/4.0/ ) 1. Introduction Oil paintings are susceptible to various alterations due to the interaction between their chemical components and the influence of external factors. One of these alterations consists on the appearance of whitish substances covering part of or the entire surface of the painting. These substances may have different morphologies and can have a number of causes. Although it is possible to find their source in biological agents such as microorganisms [ 1 , 2 ], or mechanical processes like micro-cracks on the surface, usually complex internal and physico-chemical ageing mechanisms seem to be involved in the formation process. One of them is the crys- ∗Corresponding author. E-mail address: oskar[email protected] (O. González-Mendia) . tallization of salts, namely carbonates, chlorides, acetates, oxalates, lactates or sulfates, the most frequent being magnesium sulfate heptahydrate, known as epsomite [ 3 ]. Another possible cause is the migration of free fatty acids generated by the hydrolysis of triglycerides present in the binder or in wetting and extending additives such as stearates or beeswax [ 4 , 5 ]. Among these processes, metal soap formation is regarded as the primary phenomenon associated with surface white hazes, currently generating the greatest concern in the conservation of oil paintings. In fact, the widespread presence of these soaps in both traditional and contemporary works of art not only results in substantial modifications to their aesthetic qualities but is also associated with conservation issues. For instance, the formation of aggregates within the paint layers can generate tensions and lead to delamination and loss of cohesion [ 6 ]. Metal soaps can also https://doi.org/10.1016/j.culher.2025.06.007 1296-2074/© 2025 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY-NC license ( http://creativecommons.org/licenses/by-nc/4.0/ ) E. Tarilonte, O. González-Mendia, I. Costantini et al. Journal of Cultural Heritage 74 (2025) 195–203 form large clusters, break the surface and appear in the form of protrusions [ 7 , 8 ]. The formation of metal soaps, described extensively in the literature, occurs when free fatty acids react with the metal cations of some pigments or additives, forming metal carboxylates [ 9–11 ]. In addition, agents such as high relative humidity and temperature levels, either maintained over time or with cyclic variations, seem to accentuate the process [ 12 , 13 ]. The most frequently observed metal soaps in oil paints have been lead and zinc stearates and palmitates, but carboxylates of other elements such as calcium, copper, cadmium, cobalt, iron or tin have also been identified [ 9 ]. Considering all the aforementioned, the identification of metal soaps is of vital importance for understanding their formation processes, which is essential for the proper conservation of oil painting. The identification has been usually carried out by means of spectroscopic techniques such as Fourier-Transform Infrared (FTIR) or Raman [ 14 , 15 ] and X-Ray Diffraction (XRD) [ 16 ]. Gas Chromatography-Mass Spectrometry (GC–MS) is particularly suitable for identifying the organic fraction [ 17 ]. Furthermore, elemental analysis techniques such as X-Ray Fluorescence (XRF) or Scanning Electron Microscopy-Energy Dispersive Spectroscopy (SEMEDX) are useful for obtaining information on the nature of the metal ion, as well as other inorganic substances [ 18–20 ]. SEM-EDX is especially helpful in studying the stratigraphic distribution of the soaps at greater magnification [ 20 ] and it is complementary to the information obtained by Digital Microscopy (DM) and Optical Microscopy (OM) [ 21 , 22 ]. In this work, all these analytical techniques have been combined to study a contemporary oil painting signed by the Basque artist Santos Iñurrieta (1950–2023), owned by the Orbegozo Foundation. This is one of the most significant private contemporary art collections in the Basque Country and gathers representative artworks from a key period where social and political changes facilitated the introduction of new materials and methods that now exhibit specific alterations. Untitled (1971) is an unvarnished oil on canvas that exhibits a whitish, homogeneous brittle substance that covers part of the surface, creating a haze that modifies the appearance of the painting. The multi-analytical approach included also two techniques not widely applied in paint analysis. On the one hand, X-Ray Photoelectron Spectroscopy (XPS) [ 23 , 24 ] was used to obtain surface information and, on the other hand, X-Ray Computed Tomography (CT) was employed to obtain threedimensional information of a paint fragment and improve the understanding of the internal structure [ 25 , 26 ]. 2. Research aim This work aims to be the most complete analytical study to date for the identification of surface whitening phenomena on paintings. For this objective, a comprehensive multi-analytical approach including general examination (UV induced fluorescence and radiography), different microscopic techniques (digital, optical, optical with UV, electronic), CT, SEM-EDX, μ-ED-XRF, Raman, ATR-FTIR, XRD, GC–MS and XPS was applied to the analysis of a representative contemporary oil painting covered by a white haze. 3. Material and methods 3.1. Artwork and sampling Scrapings of areas with and without white haze and microfragments were collected from the upper left red area of Untitled ( Fig. 1 ) using a scalpel and a stereomicroscope (detailed information on the number of samples and techniques applied to each is provided in Table S1). For cross-sections preparation two paint fragments were embedded in rapid cold-curing bicomponent resin Technovit® 4004 (Heraeus Kulzer GmbH), cut crosswise using a Struers Minitom and wet polished with silica abrasive paper (P40 0 0 grades), using a Struers polishing holder. The embedded samples were further polished using MicroCloth (Buehler) with polycrystalline Diamond suspension MetaDiTM Supreme 1F μm (Buehler). 3.2. Analytical methods 3.2.1. General examination and digital microscopy The painting was photographed under visible and UV light (370 nm) with a Canon EOS 750D camera. Radiography was obtained using a TX-50/100 x-ray system (Radiologia S.A.) operating at 50 kV and 50 mA for 0.3 s. The digital image capture devices were reusable photostimulable phosphor plates (35.4 ×43 cm) that, after being radiated, were scanned by means of a computer’s imaging software via the FCR Primax T2 scanner (Fujifilm). DM images were obtained with a 1.3 megapixel Dino-lite AM4917MZT instrument with a magnification range from 10 to 220x. 3.2.2. Optical microscopy Nikon Eclipse Ci-Pol light microscope equipped with DS-Fi3 digital camera was used to analyze the cross-sections under visible light and a Nikon Eclipse 50 Digital equipped with a Sight 10 camera to analyze them by ultraviolet fluorescence. 3.2.3. Scanning electron microscopy-energy dispersive X-ray spectroscopy SEM-EDX of cross-sections and paint fragment surfaces was performed on JEOL JSM 7600F scanning microscope (JEOL Ltd). The instrument offers a resolution in secondary electrons of 1.2 nm at 30 kV and 3 nm at 1 kV and a resolution of 3 nm in backscattered electrons at 15 kV. To avoid the charging of the non-conducting polymer surface, about 20 nm carbon layer was deposited onto samples in a Q150T ES coater (Quorum Technologies). Secondary electron images (SEI) and back-scattered electron (BSE) images were taken at 20 kV acceleration voltage. EDX was performed using an Aztec Ultim max detector (Oxford Instruments). Elemental mappings were obtained using a step size of 0.15–0.75 μm. 3.2.4. Computed tomography CT measurements of a paint fragment were collected on an EasyTom XL 160/230 equipment from RX Solution Company (Metrología Sariki S.A.). The double tube and detection system was configured with the nanotube with middle spot size (80KV, 150 mA) and CCD (exposure time 1.5 s and nºaverage frames 15) options. The scanning tomography without filters was selected in continuous plus reference mode after black, gain and bean alignment. The scanning time was more or less 9 h with a voxel size of 0.7 mm, 1408 projections over 360 °and 1304 slices. The scanning size was 1.0 height and 1.3 mm diameter with final geometrical magnification of 29.54. The data sets were measured and reconstructed by XAct software from RX Solution. The reconstruction volume was obtained after spot and geometry corrections and ring filter application. Scatter and beam hardening were evaluated in order to improve the final volume reconstruction tomography. 3.2.5. μ-X-ray fluorescence A High Resolution μ-ED-XRF M4 Tornado spectrometer (Bruker Nano GmbH) was used to record elemental maps of cross-sections and paint fragment surfaces. Measurements were carried out under vacuum (20 mbar), with an X-ray rhodium anode providing a voltage of 50 kV and a maximum current of 600 mA. The instrument was equipped with a silicon detector with an achievable resolution of 145 eV (Mn K αline). The individual spectra were recorded for 196 E. Tarilonte, O. González-Mendia, I. Costantini et al. Journal of Cultural Heritage 74 (2025) 195–203 Fig. 1. Untitled (1971) by Santos Iñurreta. Oil on canvas (114 ×146 cm). 5 ms with a repetition of 5 scans and a distance between the spots of 20 μm. All maps were made considering the K αline of the corresponding element, except for lead, for which the L αline was used. Control of the instrument, as well as processing and editing of the maps, were carried out with the M4 Tornado software. 3.2.6. Raman spectroscopy The Raman analyses of cross-sections were carried out by means of an InVia Renishaw confocal micro Raman spectrometer (Renishaw plc.) coupled to a DMLM Leica microscope provided with 5 ×, 20 ×, 50 ×, 50 ×(long distance) and 100 ×lenses using 532 and 785 nm excitation lasers. The lasers were set at low power (not > 1 mW at the sample) in order to avoid thermal photodecomposition or transformation. Data acquisition was carried out using the Wire 4.2 software package (Renishaw plc.). Spectra were acquired between 10 0 and 320 0 cm−1 (average 1 cm−1 of spectral resolution) and several scans were accumulated for each spectrum in order to improve the SNR (10–20 s, 5–100 accumulations). 3.2.7. Attenuated total reflectance—Fourier transform infrared spectroscopy A Nicolet Summit X (Thermo) spectrometer was used to analyze efflorescence and paint scrapings in ATR mode with Diamond crystal (two replicates of each one). Air was used as reference. Spectra were acquired for 32 scans with 4 cm−1 resolutions in a 40 0 0–40 0 cm−1 range. The spectra were collected and evaluated with Omnic software. 3.2.8. X-ray diffraction The X-ray powder diffraction patterns of efflorescence and paint scrapings (two replicates of each one) were collected by using a PHILIPS X’PERT PRO automatic diffractometer operating at 40 kV and 40 mA, in theta-theta configuration, secondary monochromator with Cu-K αradiation ( λ= 1.5418 ˚ A) and a PIXcel solid state detector (active length in 2 θ3.347 °). Data were collected from 5 to 80 °2 θ(step size = 0.026 and time per step =10 0 0s) scan speed 0.007 °/s, at RT. A fixed divergence and antiscattering slit giving a constant volume of sample illumination were used. Preliminary identification of the initial phases was evaluated using the Powder Diffraction File (PDF2) database. PANalytical X´ Pert High Score program was used for the identification procedure. 3.2.9. Gas chromatography—mass spectrometry GC–MS analysis was carried out following the method proposed by La Nasa et al . [ 17 ] which allows to analyze mixtures of free fatty acids and metal soaps. All the reagents necessary for the analysis were purchased from Sigma-Aldrich: Isooctane (for analysis), tridecanoic acid (analytical standard), N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) containing 1 % trimethylchlorosilane and 1,1,1,3,3,3-hexamethyldisilazane (HMDS). Around 500 μg of efflorescence scrapings were derivatized following two different procedures. On the one hand, 5 μL of tridecanoic acid solution in acetone (4 μg/g) were added as IS and after drying under nitrogen flow 20 μL of BSTFA (1 % TMCS) and 150 μL of isooctane were added. This mixture was heated at 78 °C for 81 min and was used for the analysis of free fatty acids and carboxilates. On the other hand, 5 μL of tridecanoic acid solution were added as IS and after drying under nitrogen flow 20 μL of HMDS and 150 μL of isooctane were added. This mixture was heated at 60 °C for 30 min and was used for the selective analysis of free fatty acids. Both mixtures were analyzed in an Agilent gas chromatograph model HP 6890 and HP 5973 mass spectrometer. Samples were injected in splitless mode at 280 °C. GC separation was performed on a fused silica capillary column HP-5MS (J&W Scientific, Agilent Technologies, stationary phase 5 % diphenyl-95 % dimethylpolysiloxane, 30 m length, 0.25 mm i.d., 0.25 mm film thickness). Chromatographic conditions were: initial temperature 80 °C, 2 min isothermal, 20 °C/min up to 280 °C, 10 min isothermal. MS parameters were: electron impact ionization (EI, 70 eV) in positive mode; ion source temperature 230 °C; interface temperature 280 °C. Chromatograms were acquired in full scan (range 50–700 m /z ) and the 197 E. Tarilonte, O. González-Mendia, I. Costantini et al. Journal of Cultural Heritage 74 (2025) 195–203 Fig. 2. Detail of Untitled (1971): Vis image (a), UV induced fluorescence (b), radiography (c). following ions were extracted: palmitic acid m/z 313, stearic acid m/z 341 and tridecanoic acid m/z 271. The injection volume was 2 μL. A Supelco 37 component FAME Mix (Sigma-Aldrich) standard was run before the analysis. 3.2.10. X-ray photoelectron spectroscopy XPS experiments were recorded with a Versaprobe III AD Physical Electronics (ULVAC) system equipped with a monochromatic AlK αradiation source (1486.7 eV). An initial analysis was carried out in order to determine the elements present in the surface of a paint fragment (wide scan: step energy 0.2 eV, pass energy 224 eV) followed by a more exhaustive analysis of the identified elements (detail scan: step energy 0.05 eV, pass energy 27 eV, time per step 20 ms) with an electron take-off angle of 45 °The spectrometer had previously been calibrated by using the Ag(3d5/2) line at 368.26 eV. The spectra were fitted with the Casa XPS 2.3.26 software, which models the contributions, after background subtraction (Shirley). 4. Results 4.1. General examination The white haze over the red paint of the artwork can be clearly observed in the detail photography in Fig. 2a . Under UV light, the efflorescence exhibits a bluish fluorescence that is even noticeable in areas where it is not visible to the naked eye ( Fig. 2b ). It is well known that zinc and lead metal soaps can exhibit heterogeneous luminescence [ 27 ]. When studying the same painting area with Xrays, no evidence of the efflorescence was observed ( Fig. 2c ). This is not surprising, since metal soaps might offer a different response depending on the thickness of the layer and the contrast with the rest of the layers of the painting. Indeed, Noble [ 9 ] points out that the aggregates can appear as either dark or light depending on the absorbance of the layer in which they are formed. 4.2. Microscopic analysis The digital microscope image shows the aspect of the whitish efflorescence ( Fig. 3a ). It is important to note that after rolling over the efflorescence up to 30 times a cotton swab dipped in deionized water, it remained unaltered, which indicates a lipophilic nature. When a closer view of the surface was obtained thanks to SEM images, a remarkable difference was observed between the haze rich area ( Fig. 3b ) and the paint surface ( Fig. 3c ). The efflorescence morphology shows irregular plates homogeneous in size (1 to 5 μm wide and around 0.1 μm thick) that resemble the shape of the efflorescence identified by Wingel et al. [ 21 ] as zinc soaps in a painting by Edwin Austin Abbey. The efflorescence could also be observed by OM in the top layer of the cross-section with a thickness of around 15 μm ( Fig. 4a ). Under UV illumination ( Fig. 4b ) the haze exhibited a fluorescence in agreement with the results obtained in the general analysis. These cross-sections images are also useful to understand the paint layer structure of this area of the artwork that seems to be formed from bottom to top by: a white preparatory layer, a yellow-orange layer (paint-layer 1), a pink layer (paint-layer 2) and a purple layer (paint-layer 3). This structure can be better observed in the crosssection of Fig. S1. The BSE image ( Fig. 4c ) shows that the preparation layer is quite homogeneous, while the paint layers are heterogeneous and rich in high atomic mass particulate substances, especially in the upper portion of the top paint layer, which could be associated with the presence of a heavy element such as lead. The surface efflorescence can be also observed in BSE and appears in a darker tone than the heavy element containing substances but brighter than the binding medium and some of the other pigment particles. That might be due to the higher organic content of the metal soaps in comparison with the heavy element containing substance [ 28 ]. 4.3. Computed tomography Tomographic reconstruction was performed based on a regiongrowing algorithm that creates a representative 3D image of the entire volume. The procedure consists of selecting a set of “seed voxels” corresponding to the material or property of interest and growing regions from these seeds [ 26 ]. In this way, an easy segmentation of materials with different absorption can be obtained. Since the aim was distinguishing metal soaps containing atoms with high atomic weight (i.e. high greyscale value), after carefully studying the greyscale histogram a threshold (isovalue of 25,154) was set in a semiautomatic way. The result obtained after filtering is shown in Fig. 5 . As in the BSE image, in the CT it can be observed how particles containing higher atomic weight elements are distributed within the paint layers. The advantage of tomography is that three-dimensional information is obtained, which can be used to better understand the distribution of these particles throughout the different paint layers. For instance, in this case a particularly interesting region of particle aggregate is identified in the top left corner of the sample that might be related to the formation of the haze. Unfortunately, at least in this scenario, the technique was not useful to distinguish the efflorescence itself, probably due to the low difference in attenuation coefficients compared with the unaltered paint surface and the fact that particles in the paint layers are heavier than the efflorescence. All the attempts made to distinguish the efflorescence using the greyscale histogram were unsuccessful. 4.4. Elemental analysis SEM-EDX allowed the identification of the elemental composition of the efflorescence, the high atomic weight particles previously detected by BSE and tomography, and the rest of the layers in the painting. Both the efflorescence and the heavy element containing particles were rich in lead, as corresponds to a system in which led soaps are formed. In Fig. 6b it can be observed that the preparatory layer does not contain lead, while all the paint layers do, being the upper layer the richest in this element. Furthermore, an area especially rich in lead can be observed in the uppermost layer of the cross-section that is associated with the efflorescence. The elemental mapping of calcium ( Fig. 6c ) and zinc ( Fig. 6d ) are also shown since they are common counter-ions in metal soaps. In this case, the presence of these ions in the efflorescence layer is very scarce, but calcium was abundant in paint-layer 1 and 2 and zinc in the preparatory layer and in the paint-layer 2. Indeed, these elemental mappings together with the mappings of other elements (Fig. S2) provided a preliminary understanding of the materials used in the painting, as it was further confirmed by Ra198 E. Tarilonte, O. González-Mendia, I. Costantini et al. Journal of Cultural Heritage 74 (2025) 195–203 Fig. 3. Image obtained with DM (x50) of the surface of a microsample of the painting rich in white efflorescence (a). SEI of an area where the white haze is present (b), and an area where the white haze is no present (c). Fig. 4. Cross-section showing the layer distribution (0: preparatory layer/1: paint-layer 1/2: paint-layer 2/3: paint-layer 3/4: whitish efflorescence) (a), under epi–illumination (b), in UV-induced fluorescence mode (c) and BSE image (d). Fig. 5. Nanotomography of a paint sample. man and XRD analysis: preparatory layer was formed by zinc and titanium white, paint-layer 1 was formed by an iron containing earth-pigment, calcium carbonate was used as a filler in the paints Fig. 6. SEM images from cross-section: BSE (a)/EDX mappings of Pb (b), Ca (c) and Zn (d). and barite could be found in both the preparation and painting layers. Elemental analysis of surface fragments was also carried out in order to obtain more information about the efflorescence and to study the distribution in a different axis. This surface analysis 199 E. Tarilonte, O. González-Mendia, I. Costantini et al. Journal of Cultural Heritage 74 (2025) 195–203 Fig. 7. Microscopic image of a microsample surface (a) and μ-ED-XRF mappings of Pb (b), Ca (c) and Zn (d). Fig. 8. Raman spectra obtained from the white haze area showing a characteristic spectrum of metal soaps (a) and lead containing particles found in the paint layers identified as litharge (b). The bands marked with R correspond to the embedding resin (785 nm laser excitation). was performed by SEM-EDX and μ-ED-XRF and, although the former offers a better spatial resolution, the images obtained with the later allowed a suitable interpretation in a faster way without requiring the coating of the sample. In the μ-ED-XRF maps in Fig. 7 , it can be observed that the efflorescence is rich in lead ( Fig. 7b ) and that the mapping of this element matches to a large extent the distribution of the haze. Again, no significant presence of calcium ( Fig. 7c ) or zinc ( Fig. 7d ) can be detected in the efflorescence. Furthermore, these mappings were useful to identify areas of the painting that were not affected by this phenomenon. 4.5. Raman analysis The Raman spectrum obtained from the uppermost layer of the cross-section ( Fig. 8a ) exhibited the characteristic bands of metal soaps (Raman bands at 890, 925, 1064, 1002, 1131, 1296, 1416, and 1440 cm−1 ) plus signals from the resin in which the sample was embedded (Raman bands at 360, 482, 600, and 812 cm−1 ). The characteristic bands could be attributed mainly to palmitates or stearates of a wide range of metallic ions [ 15 , 29 , 30 ]. Furthermore, Raman spectroscopy allowed us to identify the lead containing particles found in the paint layers as lead (II) oxide litharge ( αPbO, main Raman bands at 147 cm−1 ) ( Fig. 8b ). Although the most common source of lead soaps are pigments such as lead white, chrome yellow or lead and tin yellow [ 31 ], none of these pigments was identified in the samples in addition to litharge. Nevertheless, the formation of lead soaps has also been attributed to other lead containing substances such as oxides used as driers [ 32 ]. Hence, the most probable source of lead involved in the formation of the metal soap is a lead-based drying agent. Besides the information regarding the efflorescence and the lead containing substance, Raman images (obtained considering the band with the highest intensity for each compound) offered a more throughout view of the painting technique used by Santos Iñurreta. In this way, pigments used in the paint layers were identified as α-goethite ( α-FeO(OH), Raman band 385 cm−1 ) in paintlayer 1, dioxazine violet (PV23, Raman band 1205 cm−1 ) in paintlayer 2, a mixture of naphthol red AS (PR146, Raman band 1582 cm−1 ) and phthalo green (PG7, Raman band 748 cm−1 ) in paintlayer 3. Furthermore the pigments and fillers inferred from elemental analysis were confirmed: barite (BaSO4 , Raman band 987 cm−1 ), calcite (CaCO3 , Raman band 1086 cm−1 ), quartz (SiO2 , Raman band 465 cm−1 ) and titanium oxide, mainly rutile (TiO2 , Raman band 612 cm−1 ) with traces of anatase (TiO2 , Raman band 142 cm−1 ). See Fig. S3 for more information. 4.6. Analysis of the efflorescence by ATR-FTIR The ATR-FTIR spectra obtained from the efflorescence and paint scrapings exhibited a marked difference in the range 180 0–120 0 cm−1 ( Fig. 9 ). Indeed, this is the area in which stretching profiles for oil paint and metal soaps are expected to differ [ 29 ]. Unlike Raman spectroscopy, FTIR allows to distinguish the metal cations based on the COO−asymmetric stretching band between 1600 and 1500 cm−1 while the length of the carbon chain is more difficult to elucidate. According to the bands in 1538 and 1511 cm−1 and the reference spectra available in the literature [ 10 , 14 , 29 ] the efflorescence seems to be formed mainly by lead palmitate and/or stearate. Discriminating between these species is a difficult task since the only difference is the series of CH2 progression bands between 1350 and 1175 cm−1 where stearate shows an additional band [ 14 , 33 ]. The bands observed in that region for the efflorescence are complex and suggest that the soap is not formed by a single species but by a mixture of stearate and palmitate, a phenomenon already observed by other authors [ 33 ]. 4.7. Analysis of the efflorescence by XRD As expected, diffraction patterns of painting and white haze scrapings (Fig. S4) showed many similarities since sampling of the efflorescence inevitably involved also the sampling of particles of the surrounding painting. Nevertheless, there were significant differences, especially at low angles. In this way, some of the differences could be attributed to palmitic and stearic free fatty acids, but the most remarkable differences fall within the range of lead soaps. However, the D-spacings did not match either with lead palmitate or with lead stearate, but lied between them ( Fig. 10 ). These intermediate D-spacings were also observed by Salvadó et al. [ 33 ] and attributed to the formation of a mixture of lead palmitate and lead stearate. Indeed, these results agree with the observed by ATR-FTIR. 200 E. Tarilonte, O. González-Mendia, I. Costantini et al. Journal of Cultural Heritage 74 (2025) 195–203 Fig. 9. ATR-FTIR spectra of scrapings from an area rich in white haze (red) and from a paint sample without white haze (blue). Signals at 1538 and 1511 cm−1 correspond to the asymmetric COO−stretching bands of a lead soap. The black square shows a detail of the CH2 progression bands of the aliphatic chain. Fig. 10. Magnification of the diffraction pattern at low angles overlapped with the expected signals for lead palmitate (blue) and lead stearate (green). XRD analysis also confirmed the presence of some compounds in the painting, previously identified by Raman spectroscopy: goethite, dioxazine violet, naphthol red AS, barite, calcite, rutile, zinc oxide, quartz and lead oxide (Fig. S4). Lead oxide showed a tetragonal crystal system confirming the presence of litharge, which again suggests its use as drier in the oil paint [ 32 ]. 4.8. Analysis of the efflorescence by GC–MS GC–MS analysis was carried out in order to obtain information about the organic fraction of the efflorescence. On the one hand, when free fatty acids and carboxylates were analyzed together, the most abundant species were palmitic and stearic acids followed, to a lesser extent, by azelaic and suberic acids. This is in fact the composition expected for a cured oil paint film [ 34 ]. On the other hand, when free acid and carboxylate forms were distinguished, the latter ones proved to be predominant (3.1 ratio for palmitate/palmitic acid and 4.1 ratio for stearate/stearic acid, see Fig. S5). These findings indicate that the efflorescence is mainly a mixture of lead palmitate and lead stearate, with minor amounts of the corresponding free fatty acids, which also agrees with the results obtained by FTIR and XRD. 4.9. Analysis of the efflorescence by XPS The XPS analysis revealed the presence of carbon, oxygen and lead in a paint fragment surface partially covered by the white haze (Table S2). Although the amount of lead was higher when examining areas rich in haze, it is important to note that this metal was also detected in apparently haze free areas. Although XPS is not a technique of choice for the characterization of metal soaps, we obtained some results that deserve attention and should be further investigated. Firstly, the binding energies for oxygen suffered a significant decrease in the efflorescence (Fig. S6). This might be explained by the fact that oxygen is in a carboxylate coordinated with a lead (II) ion while in oil painting is mainly in ester or free acid form. Even if metal soaps have not been widely studied by XPS, these results could be compared with those ones obtained by Carvalho [ 24 ] for cupper palmitate. Secondly, and more surprisingly, the signal of lead in the efflorescence showed a significant contribution of Pb(0) (Fig. S7). In the extremely unlikely case that metallic lead was present in the paint surface it should have been detected by XRD, so this finding is probably an artefact of the XPS analysis. Indeed other authors have reported Pb(0) formation in paint analysis under high-intensity X-ray beams [ 35 ] or in other lead containing compounds during XPS analysis [ 36 ]. This artefact was only detected in the efflorescence analysis and not in the efflorescence free painting, therefore, lead soaps might be prone to photolysis under these analytical conditions. 5. Conclusions This study demonstrates the efficacy of employing a multianalytical approach to characterize and understand surface whitening phenomena formation in oil paintings. Furthermore, it provides a detailed insight into the utility of each technique and the specific type of information it offers for a study case in which metal soaps were identified. Elemental analysis by SEM-EDX and μ-EDXRF facilitated the study of the distribution of lead ions in crosssections and paint fragment surfaces, and showed that lead derives from paint layers. Regarding the characterization of the metal soap, XRD, Raman and ATR-FTIR proved to be useful and complementary techniques. ATR-FTIR should be especially addressed since, using 201 E. Tarilonte, O. González-Mendia, I. Costantini et al. Journal of Cultural Heritage 74 (2025) 195–203 a relative simple instrumentation, both the metal ion and the organic chain could be elucidated, while Raman spectroscopy reveals constrains in identifying the counter-ion. Nevertheless, the characterization of the organic fraction is not always feasible by ATRFTIR when dealing with complex samples as it is the case study in which signals of different compounds can overlap. In those situations, the application of GC–MS demonstrated to be a valid strategy, especially considering that carboxylates and free acid forms can be distinguished following a two-step derivatization procedure. XPS analysis did not offer additional information about metal soaps but the shift observed in binding energies when comparing the efflorescence and the oil paint spectra opens new research possibilities. Similarly, CT image obtained from the analyzed paint fragment did not reflect the distribution of the haze but was useful to trace particles containing atoms with high atomic weight in threedimensions. It is important to highlight that even if all the point analysis techniques require sampling, only GC–MS is destructive, and therefore, with a carefully planned analytical workflow, sample collection can be minimized while maximizing analytical outcomes. With respect to the metal soap formation process in Santos Iñurreta’s Untitled (1971), information obtained from SEM-EDX and Raman images suggest that the lead source is litharge used as a drier and not a lead pigment. Lead might have interacted with stearic and palmitic acids from oil to create lead stearate and palmitate. The manual addition of the drier could have contributed to an uneven distribution of the whitish haze. The multi-analytical approach also revealed the layer structure of the artwork formed by a preparation layer rich in zinc and titanium white and three layers containing, from bottom to top, goethite, dioxazine violet and a mixture of phthalo green and naphthol red AS, together with fillers such as calcium carbonate, barite or quartz. In this way, this work offers valuable insights into the artist’s technique and, to the best of our knowledge; it is the first study of this kind conducted on a painting by Santos Iñurrieta. The strategy applied in this study is transferable to the analysis and conservation of oil paintings experiencing other degradation phenomena such as salt formation or fatty acids efflorescence. In fact, due to the flexibility of the multi-analytical framework, techniques that do not offer conclusive results for metal soaps might offer more valuable information for other whitening phenomena as in the case of XPS for salt analysis or GC–MS for fatty acid determination. It is necessary to bear in mind that all these surface whitening substances not only affect the aesthetics of the artworks, but also represent a conservation problem. Despite recent advances in this field, much remains to be investigated about the underlying causes and the mechanisms that generate them. In this way, this work contributes to the development of better conservation strategies for preserving oil-paintings and the consequent research for effective solutions for the long-term preservation of art. Acknowledgments This work was supported by funding from the Basque Government to the FARMARTEM consolidated group (grant number 1673/22 ). Also, we would thank for technical and human support (Luis Bartolomé, Sergio Fernández, Aitor Larrañaga, Belén Sánchez and Leire Sanfelices) provided by SGIker of UPV/EHU and European funding (ERDF and ESF). Likewise, we are grateful to the Orbegozo Foundation for providing the artwork for study. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.culher.2025.06.007 . References [1] C.L. V˘ acar, C. Mircea, M. Parvu, M.D. Podar, Diversity and metabolic activity of fungi causing biodeterioration of canvas paintings, J. Fungi 8 (6) (2022) 589, doi: 10.3390/jof8060589 . [2] C.N. Zmeu, P. Bosch-Roig, Risk analysis of biodeterioration in contemporary art collections: the poly-material challenge, J. Cult. Herit. 58 (2022) 33–48, doi: 10. 1016/j.culher.2022.09.014 . [3] J. Harrison, J. Lee, B. Ormsby, D.J. Payne, The influence of light and relative humidity on the formation of epsomite in cadmium yellow and French ultramarine modern oil paints, Herit. Sci. 9 (1) (2021) 107, doi: 10.1186/ s40494021005692 . [4] K. Keune, E. Ferreira, J.J. Boon, Characterization and localization of the oilbinding medium in paint cross-sections using imaging secondary ion mass spectrometry, in: ICOM Committee for Conservation 14th Triennial Meeting The Hague 12-16 September 2 005. The Hague, James & James/Earthscan, 2005, pp. 796–802. https://www.icomccpublicationsonline.org/2128/ Characterizationandlocalizationoftheoilbindingmediuminpaintcrosssectionsusingimagingsecondaryionmassspectrometry . [5] F.C. Izzo, K.J. van den Berg, H. van Heulen, B. Ferriani, E. Zendri, Modern oil paints—formulations, organic additives and degradation: some case studies, in: K.J. van den Berg, A. Burnstock, M. de Keijzer, J. Krueger, T. Learner, A. Tagle, G. Heydenreich (Eds.), Issues in contemporary oil paint, Springer, Cham, Switzerland, 2014, pp. 75–104, doi: 10.1007/9783319101002_5 . [6] Y. Maor, A. Murray, Delamination of oil paints from acrylic grounds, MRS Online Proc. Libr. 1047 (2008) 401, doi: 10.1557/PROC1047Y0401 . [7] K. Helwig, J. Poulin, M.C. Corbeil, E. Moffatt, D. Duguay, Conservation issues in several twentieth-century Canadian oil paintings: the role of zinc carboxylate reaction products, in: K.J. van den Berg, A. Burnstock, M. de Keijzer, J. Krueger, T. Learner, A. Tagle, G. Heydenreich (Eds.), Issues in contemporary oil paint, Springer, Cham, Switzerland, 2014, pp. 167–184, doi: 10.1007/ 978331910100-2_11 . [8] I.A. Tank Bronken, J.J. Boon, R. Corkery, C.C. Steindal, Changing surface features, weeping and metal soap formation in paintings by Karel Appel and Asger Jorn from 1946 to 1971, J. Cult. Herit. 35 (2019) 279–287, doi: 10.1016/j.culher.2018. 09.005 . [9] P. Noble, A brief history of metal soaps in paintings from a conservation perspective, in: F. Casadio, K. Keune, P. Noble, A. Van Loon, E. Hendriks, S.A. Centeno, G. Osmond (Eds.), Metal soaps in art: conservation and research, Springer, Cham, Switzerland, 2019, pp. 1–22, doi: 10.1007/978331990617-1_1 . [10] J.J. Hermans, Metal soaps in oil paint: structure, mechanisms and dynamics (2017) PhD thesis https://dare.uva.nl/search?identifier=53663926-183c40aab7b3e6027979cb7d . [11] A . Sawicka, A . Burnstock, F.C. Izzo, K. Keune, J.J. Boon, K. Krisch, K.J. van den Berg, An investigation into the viability of removal of lead soap efflorescence from contemporary oil paintings, in: K.J. van den Berg, A. Burnstock, M. de Keijzer, J. Krueger, T. Learner, A. Tagle, G. Heydenreich (Eds.), Issues in contemporary oil paint, Springer, Cham, Switzerland, 2014, pp. 311–332, doi: 10.1007/978331910100-2_21 . [12] C. Higgitt, M. Spring, D. Saunders, Pigment-medium interactions in oil paint films containing red lead or lead-tin yellow, Natl. Gallery Tech. Bull. 24 (2003) 75–95 https://www.nationalgallery.org.uk/media/15504/higgitt_spring_ saunders2003.pdf . [13] J.J. Boon, F. Hoogland, K. Keune, Chemical processes in aged oil paints affecting metal soap migration and aggregation, in: 34th annual meeting of the AIC of Historic & Artistic Works providence. Rhode Island, June 16-19. Rhode Island, H. Mar Parkin, AIC, 2007, pp. 16–23. https://amolf.nl/publications/ chemicalprocessesinagedoilpaintsaffectingmetalsoapmigrationandaggregation . [14] L. Robinet, M.C. Corbeil, The characterization of metal soaps, Stud. Conserv. 48 (1) (2003) 23–40, doi: 10.1179/sic.2003.48.1.23 . [15] F.C. Izzo, M. Kratter, A. Nevin, E. Zendri, A critical review on the analysis of metal soaps in oil paintings, ChemistryOpen 10 (9) (2021) 904–921, doi: 10. 1002/open.202100166 . [16] S. Garrappa, D. Hradil, J. Hradilová, E. Koˇ cí, M. Pech, P. Bezdiˇ cka, S. Švarcová, Non-invasive identification of lead soaps in painted miniatures, Anal. Bioanal. Chem. 413 (1) (2021) 263–278, doi: 10.10 07/s0 0216020029987 . [17] J. La Nasa, F. Modugno, M. Aloisi, A. Lluveras-Tenorio, I. Bonaduce, Development of a GC/MS method for the qualitative and quantitative analysis of mixtures of free fatty acids and metal soaps in paint samples, Anal. Chim. Acta 1001 (2018) 51–58, doi: 10.1016/j.aca.2017.11.017 . [18] V. Otero, M. Vilarigues, L. Carlyle, M. Cotte, W. De Nolf, M.J. Melo, A little key to oxalate formation in oil paints: protective patina or chemical reactor? Photochem. Photobiol. Sci. 17 (3) (2018) 266–270, doi: 10.1039/c7pp00307b . [19] W. Faubel, R. Simon, S. Heissler, F. Friedrich, P.G. Weidler, H. Becker, W. Schmidt, Protrusions in a painting by Max Beckmann examined with confocal μ-XRF, J. Anal. Spectrom. 26 (5) (2011) 942–948, doi: 10.1039/ C0JA00178C . [20] C. Romano, T. Lam, G.A. Newsome, J.A. Taillon, N. Little, J. Tsang, Characterization of zinc carboxylates in an oil paint test panel, Stud. Conserv. 65 (1) (2020) 14–27, doi: 10.1080/00393630.2019.1666467 . [21] K. Wingel, R. Hark, K. Schilling, C. Schwarz, A. Bezur, Efflorescence on the paintings of Edwin Austin Abbey: examination, analysis, and cleaning of surface bloom on the spirit of light, in: K. van den Berg, I. Bonaduce, A. Burnstock, B. Ormsby, M. Scharff, L. Carlyle, G. Heydenreich, K. Keune (Eds.), Conserva202 E. Tarilonte, O. González-Mendia, I. Costantini et al. Journal of Cultural Heritage 74 (2025) 195–203 tion of modern oil paintings, Springer, Cham, Switzerland, 2019, pp. 297–307, doi: 10.1007/9783030192549_23 . [22] A. Graczyk, P. Hélou-de La Grandière, A. Phenix, S. Mirabaud, Oil paint straight from the tube: paint-specific deterioration in works by Alexis Mérodack-Jeaneau, 1910–1913 In: van den Berg, in: KI. Bonaduce, A. Burnstock, B. Ormsby, M. Scharff, L. Carlyle, G. Heydenreich, K. Keune (Eds.), Conservation of modern oil paintings, Springer, Cham, Switzerland, 2019, pp. 229–243, doi: 10.1007/9783030192549_17 . [23] J. Lee, B. Ormsby, A. Burnstock, K.J. van den Berg, Modern oil paintings in Tate’s collection: a review of analytical findings and reflections on water-sensitivity, in: K.J. van den Berg, I. Bonaduce, A. Burnstock, B. Ormsby, M. Scharff, L. Carlyle, G. Heydenreich, K. Keune (Eds.), Conservation of modern oil paintings, Springer International Publishing, Cham, 2019, pp. 495–522, doi: 10.1007/ 9783030192549_38 . [24] L. Carvahlo, Cooper-organic complexes synthetized electrochemically, in: Beyond copper soaps. SpringerBriefs in applied sciences and technology, Springer International Publishing, Cham, 2022, pp. 13–39, doi: 10.1007/ 9783030978921_3 . [25] F.T.H. Broers, I. Verslype, K.W. Bossers, F. Vanmeert, V. Gonzalez, J. Garrevoet, A. van Loon, E. van Duijn, A. Krekeler, N. de Keyser, I. Steeman, P. Noble, K. Janssens, F. Meirer, K. Keune, Correlated x-ray fluorescence and ptychographic nano-tomography on Rembrandt’s the Night Watch reveals unknown lead “layer, Sci. Adv 9 (50) (2023) eadj9394, doi: 10.1126/sciadv.adj9394 . [26] C. Gervais, J.J. Boon, F. Marone, E.S.B. Ferreira, Characterization of porosity in a 19th century painting ground by synchrotron radiation X-ray tomography, Appl. Phys. A 111 (2013) 31–38, doi: 10.10 07/s0 03390127533y . [27] M. Thoury, A. Van Loon, K. Keune, J.J. Hermans, M. Réfrégiers, Photoluminescence micro-imaging sheds new light on the development of metal soaps in oil paintings, in: F. Casadio, K. Keune, P. Noble, A. Van Loon, E. Hendriks, S.A. Centeno, G. Osmond (Eds.), Metal soaps in art: conservation and research, Springer, Cham, Switzerland, 2019, pp. 211–225. https://doi.org/10. 1007/9783319906171_12 . [28] K. Keune, R.P. Kramer, S. Stangier, M.H. van Eikema Hommes, Impact of lead soaps on the formation of age craquelure, in: F. Casadio, K. Keune, P. Noble, A. Van Loon, E. Hendriks, S.A. Centeno, G. Osmond (Eds.), Metal soaps in art: conservation and research, Springer, Cham, Switzerland, 2019, pp. 107–121, doi: 10.1007/9783319906171_6 . [29] V. Otero, D. Sanches, C. Montagner, M. Vilarigues, L. Carlyle, J.A. Lopes, M.J. Melo, Characterisation of metal carboxylates by Raman and infrared spectroscopy in works of art, Raman Art. Archaeol. 45 (11–12) (2014) 1197–1206, doi: 10.1002/jrs.4520 . [30] E. Platania, N.L.W. Streeton, A. Vila, D. Buti, F. Caruso, E. Uggerud, Investigation of mineralization products of lead soaps in a late medieval panel painting, Spectrochim. Acta A: Mol. Biomol. Spectrosc. 228 (2020) 117844, doi: 10.1016/j.saa.2019.117844 . [31] C.S. Tumosa, M.F. Mecklenburg, The influence of lead ions on the drying of oils, Stud. Conserv. 50 (sup1) (2005) 39–47, doi: 10.1179/sic.2005.50.Supplement-1. 39 . [32] M. Cotte, E. Checroun, W. De Nolf, Y. Taniguchi, L. De Viguerie, M. Burghammer, P. Walter, C. Rivard, M. Salomé, K. Janssens, J. Susini, Lead soaps in paintings: friends or foes? Stud. Conserv. 62 (1) (2016) 2–23, doi: 10.1080/00393630.2016. 1232529 . [33] N. Salvadó, S. Butí, T. Pradell, V. Beltran, G. Cinque, J. Juanhuix, Identification and distribution of metal soaps and oxalates in oil and tempera paint layers in fifteenth-century altarpieces using synchrotron radiation techniques, in: F. Casadio, K. Keune, P. Noble, A. Van Loon, E. Hendriks, S.A. Centeno, G. Osmond (Eds.), Metal soaps in art: conservation and research, Springer, Cham, Switzerland, 2019, pp. 195–210, doi: 10.1007/9783319906171_11 . [34] G. Osmond, G. Zinc soaps: an overview of zinc oxide reactivity and consequences of soap formation in oil-based paintings, in: F. Casadio, K. Keune, P. Noble, A. Van Loon, E. Hendriks, S.A. Centeno, G. Osmond (Eds.), Metal soaps in art: conservation and research, Springer, Cham, Switzerland, 2019, pp. 25– 46, doi: 10.1007/9783319906171_2 . [35] V. Gonzalez, I. Fazlic, M. Cotte, F. Vanmeert, A. Gestels, S. De Meyer, F. Broers, J. Hermans, A. van Loon, K. Janssens, P. Noble, K. Keune, Lead(II) formate in Rembrandt’s night watch: detection and distribution from the macroto the micro-scale, Angew. Chem. Int. Ed. 62 (16) (2023) e202216478, doi: 10.1002/ anie.202216478 . [36] J.D. McGettrick, K. Hooper, A. Pockett, J. Baker, J. Troughton, M. Carnie, T. Watson, Sources of Pb(0) artefacts during XPS analysis of lead halide perovskites, Mater. Lett 251 (2019) 98–101, doi: 10.1016/j.matlet.2019.04.081 . 203